Package com.microsoft.semantickernel.aiservices.openai.textembedding

Class OpenAITextEmbeddingGenerationService

java.lang.Object

com.microsoft.semantickernel.aiservices.openai.OpenAiService<com.azure.ai.openai.OpenAIAsyncClient>

com.microsoft.semantickernel.aiservices.openai.textembedding.OpenAITextEmbeddingGenerationService

All Implemented Interfaces:

AIService, EmbeddingGenerationService<String>, TextEmbeddingGenerationService

public class OpenAITextEmbeddingGenerationService
extends OpenAiService<com.azure.ai.openai.OpenAIAsyncClient>
implements TextEmbeddingGenerationService

An OpenAI implementation of a TextEmbeddingGenerationService.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	OpenAITextEmbeddingGenerationService.Builder	A builder for creating a OpenAITextEmbeddingGenerationService.

Field Summary

Fields

Modifier and Type	Field	Description
static int	EMBEDDING_DIMENSIONS_LARGE	Dimension of the OpenAI text-embedding-3-large model.
static int	EMBEDDING_DIMENSIONS_SMALL	Dimension of the OpenAI text-embedding-3-small model.

Constructor Summary

Constructors

Constructor	Description
OpenAITextEmbeddingGenerationService(com.azure.ai.openai.OpenAIAsyncClient client, String deploymentName, String modelId, String serviceId, int dimensions)	Creates a new OpenAITextEmbeddingGenerationService.

Method Summary

Modifier and Type	Method	Description
static OpenAITextEmbeddingGenerationService.Builder	builder()	Creates a builder for creating a OpenAITextEmbeddingGenerationService.
reactor.core.publisher.Mono<Embedding>	generateEmbeddingAsync(String data)	Generates embeddings for the given data.
reactor.core.publisher.Mono<List<Embedding>>	generateEmbeddingsAsync(List<String> data)	Generates embeddings for the given data.

Methods inherited from class com.microsoft.semantickernel.aiservices.openai.OpenAiService

getDeploymentName, getModelId, getServiceId

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface com.microsoft.semantickernel.services.AIService

getModelId, getServiceId

Field Detail

EMBEDDING_DIMENSIONS_SMALL

public static final int EMBEDDING_DIMENSIONS_SMALL

Dimension of the OpenAI text-embedding-3-small model.

See Also:

Constant Field Values

EMBEDDING_DIMENSIONS_LARGE

public static final int EMBEDDING_DIMENSIONS_LARGE

Dimension of the OpenAI text-embedding-3-large model.

See Also:

Constant Field Values

Constructor Detail

OpenAITextEmbeddingGenerationService

public OpenAITextEmbeddingGenerationService(com.azure.ai.openai.OpenAIAsyncClient client,
                                            String deploymentName,
                                            String modelId,
                                            @Nullable
                                            String serviceId,
                                            int dimensions)

Creates a new OpenAITextEmbeddingGenerationService.

Parameters:

client - OpenAI client

deploymentName - deployment name

dimensions - The dimensions for the embeddings.

modelId - OpenAI model id

serviceId - Service id

Method Detail

builder

public static OpenAITextEmbeddingGenerationService.Builder builder()

Creates a builder for creating a OpenAITextEmbeddingGenerationService.

Returns:

A new OpenAITextEmbeddingGenerationService builder.

generateEmbeddingAsync

public reactor.core.publisher.Mono<Embedding> generateEmbeddingAsync(String data)

Generates embeddings for the given data.

Specified by:

generateEmbeddingAsync in interface EmbeddingGenerationService<String>

Parameters:

data - The data to generate embeddings for.

Returns:

A Mono that completes with the embeddings.

generateEmbeddingsAsync

public reactor.core.publisher.Mono<List<Embedding>> generateEmbeddingsAsync(List<String> data)

Generates embeddings for the given data.

Specified by:

generateEmbeddingsAsync in interface EmbeddingGenerationService<String>

Parameters:

data - The data to generate embeddings for.

Returns:

A Mono that completes with the embeddings.